OpenVINO-YOLO-Automatic-Generation/parse_config.py at master · TNTWEN/OpenVINO-YOLO-Automatic-Generation · GitHub

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
import os
import numpy as np
import argparse


def parse_model_cfg(path):
    if not path.endswith('.cfg'):
        path += '.cfg'
    if not os.path.exists(path) and os.path.exists('cfg' + os.sep + path):
        path = 'cfg' + os.sep + path

    with open(path, 'r') as f:
        lines = f.read().split('\n')

    lines=[x for x in lines if x and not x.startswith("#")]
    lines=[x.rstrip().lstrip() for x in lines]
    mdefs = []
    for line in lines:
        if line.startswith('['):
            mdefs.append({})
            mdefs[-1]['type'] = line[1:-1].rstrip()
            if mdefs[-1]['type'] == 'convolutional':
                mdefs[-1]['batch_normalize'] = 0
        else:
            key, val = line.split("=")
            key = key.rstrip()

            if key == 'anchors':  # return nparray
                mdefs[-1][key] = np.array([float(x) for x in val.split(',')]).reshape((-1, 2))
            elif (key in ['from', 'layers', 'mask']) or (key == 'size' and ',' in val):
                mdefs[-1][key] = [int(x) for x in val.split(',')]
            else:
                val = val.strip()
                if val.isnumeric():
                    mdefs[-1][key] = int(val) if (int(val) - float(val)) == 0 else float(val)
                else:
                    mdefs[-1][key] = val

    # Check all fields are supported
    supported = ['type', 'batch_normalize', 'filters', 'size', 'stride', 'pad', 'activation', 'layers', 'groups',
                 'from', 'mask', 'anchors', 'classes', 'num', 'jitter', 'ignore_thresh', 'truth_thresh', 'random',
                 'stride_x', 'stride_y', 'weights_type', 'weights_normalization', 'scale_x_y', 'beta_nms', 'nms_kind',
                 'iou_loss', 'iou_normalizer', 'cls_normalizer', 'iou_thresh']

    f = []  # fields
    for x in mdefs[1:]:
        [f.append(k) for k in x if k not in f]
    u = [x for x in f if x not in supported]
    assert not any(u), "Unsupported fields %s in %s." % (u, path)

    return mdefs


def generate(mdefs):
    yololayer=0
    for i,mdef in enumerate(mdefs):
        if mdef['type']=='shortcut':
            _from=mdef['from'][0]
            _fromid=i+_from
            mdefs[_fromid]['extra']="shortcut=inputs"

        if mdef['type']=='route':
            for x in mdef['layers']:
                if mdefs[i+x if x<0 else x]['type']=='maxpool':
                    pass
                else :
                    if 'extra' not in mdef.keys():
                        mdefs[i+x if x<0 else x]['extra']="route%s=inputs"%str(i+x if x<0 else x)
                        mdefs[i+x if x <0 else x]['route_name']="route%s"%str(i+x if x<0 else x)


        if mdef['type']=='maxpool':
            mdef['route_name']="maxpool%s"%str(i)
            mdef['extra']="maxpool%s=slim.max_pool2d(inputs, %s, 1, 'SAME')"%(str(i),mdef['size'])

        if mdef['type']=="yolo":
            yololayer+=1
            mdef['yoloid']=yololayer


    for i, mdef in enumerate(mdefs):
        if mdef['type']=="convolutional":
            if mdef['batch_normalize']:
                print("inputs = _conv2d_fixed_padding(inputs, %s, %s,strides=%s)"%(mdef['filters'],mdef['size'],mdef['stride']))
                if 'extra' in mdef.keys():
                    print(mdef['extra'])
        if mdef['type']=="shortcut":
            print("inputs = inputs + shortcut")
            if 'extra'in mdef.keys():
                print(mdef['extra'])
        if mdef['type']=="maxpool":
            print(mdef['extra'])
        if mdef['type']=='route':
            if len(mdef['layers'])==1:
                print("inputs=%s"%mdefs[i+mdef["layers"][0] if mdef["layers"][0]<0 else mdef["layers"][0]]['route_name'])
            if len(mdef['layers'])==4:
                print("inputs=tf.concat([%s,%s,%s,%s],axis=1 if data_format == 'NCHW' else 3)"%(mdefs[i+mdef["layers"][0] if mdef["layers"][0]<0 else mdef["layers"][0]]['route_name'],\
                                                       mdefs[i+mdef["layers"][1] if mdef["layers"][1]<0 else mdef["layers"][1]]['route_name'],\
                                                       mdefs[i+mdef["layers"][2] if mdef["layers"][2]<0 else mdef["layers"][2]]['route_name'],\
                                                       mdefs[i+mdef["layers"][3] if mdef["layers"][3]<0 else mdef["layers"][3]]['route_name']))
            if len(mdef['layers'])==2:
                print("inputs=tf.concat([%s,%s],axis=1 if data_format == 'NCHW' else 3)"%(mdefs[i+mdef["layers"][0] if mdef["layers"][0]<0 else mdef["layers"][0]]['route_name'],\
                                                 mdefs[i+mdef["layers"][1] if mdef["layers"][1]<0 else mdef["layers"][1]]['route_name']))

        if mdef['type']=="upsample":
            print("inputs = _upsample(inputs,route%s.get_shape().as_list(),data_format)"%mdefs[i+1]["layers"][1])
            if 'extra'in mdef.keys():
                print(mdef['extra'])
        if mdef['type']=="yolo":
            print("detect_%s = _detection_layer(inputs, num_classes, _ANCHORS[%s:%s], img_size, data_format)\ndetect_%s = tf.identity(detect_%s, name='detect_%s')"%(mdef['yoloid'],str(mdef['mask'][0]),str(mdef['mask'][2]+1),mdef['yoloid'],mdef['yoloid'],mdef['yoloid']))


if __name__=="__main__":
    parser = argparse.ArgumentParser()
    parser.add_argument("--cfg",type=str,default='cfg/yolov3.cfg',help=('*.cfg path'))
    opt = parser.parse_args()
    path = opt.cfg
    mdefs=parse_model_cfg(path)
    mdefs.pop(0)
    generate(mdefs)
    print("detections = tf.concat([detect_1, detect_2, detect_3], axis=1)")
    print("detections = tf.identity(detections, name='detections')")